How do I install the Critique skill?

Run npx skillmds@latest add zio/critique in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Critique safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Critique?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Critique free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Critique?

zio (@zio) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Critique by zio — Agent Skill

Name: Critique
Author: zio

Critique

A coordinator-only critique-fix loop for driving an existing documentation page to approval — a fresh critic delegate reports severity-tagged findings, a fresh fixer delegate resolves them, capped at three rounds with HIGH findings iterating longest and MEDIUM getting one chance. Load when reviewing an already-written page against source, siblings, and consistency, not when writing new content.

zio edb6994 2 files · 6.0 KB Updated Sep 17, 2026

File contents

Documentation Critique Loop

The full procedure is provided verbatim in references/review-loop.md: role separation (coordinator never edits directly), how to gather critic context from paths rather than content, the critic's required findings/verdict format, the severity-based iteration rules, and multi-file handling.

Originally ported from a docs-critique plugin skill (since removed from plugins/documentation as unneeded — the pattern here stands on its own and doesn't depend on that file existing), translated from Claude Code's Agent(...) tool calls to Flue's own delegation primitive — a coordinator that mounts this skill delegates to a critic or fixer with the built-in task tool, the same way every subagent delegation in this project works, rather than a named role under src/subagents/.

Not currently mounted by any flowrite agent. This is reference material only: the pattern is documented and ready, but no standalone entry point or KINDS row wires it into a run. A future coordinator that wants a critique-fix loop over an already-written page can mount this skill; nothing here runs on its own.

Frequently asked questions